Vulnerability of melanic Daphnia to brown trout predation

(1)

Journal of Plankton Research Vol.18 no.ll pp.2113-2118, 1996

Vulnerability of melanic Daphnia to brown trout predation

Harald Saegrov¹⁴, Anders Hobaek² and Jan Henning L'Ab6e-Lund³ 'Zoological Institute, Department of Animal Ecology, University of Bergen, Allegt. 41, N-5007 Bergen, ^Norwegian Institute for Water Research, Regional

Office Bergen, Thormoehlensgt. 55, N-5008 Bergen and³Norwegian Water Resources and Energy Administration, PO Box 5091 Majorstua, N-0301 Oslo, Norway

4Present address: Radgivende Biologer A/S, Bredsgarden, Bryggen, N-5003 Bergen, Norway

Abstract The coexistence of melanic Daphnia cf. longispina and facultatively planktivorous brown trout is reported from a clear-water, alpine lake. This co-occurrence is uncommon, presumably due to the vulnerability of pigmented Daphnia to fish predation. Lake Bjornesfjorden (Norway) provided an opportunity to test this assumption. About 20% of the fish caught in gill nets had fed on Daphnia.

The trout exerted a marked selection for large-sized Daphnia prey, and a very strong selection for pig- mented individuals relative to transparent ones. The persistence of a pigmented Daphnia population probably relies on limited recruitment and a low stock of the predator, and the availability of more favourable benthic prey organisms.

Introduction

Cuticular pigments in Daphnia are known to provide protection from harmful UV radiation (Hebert and Emery, 1990; Hessen and S0rensen, 1990; Hobaek and Wolf, 1991; Hessen, 1993,1994). These melanin pigments are deposited in the dor- sally directed parts of the exoskeleton. In contrast to internal pigments like carotenoids, cuticular pigments are lost with every moult and have to be replen- ished in the newly formed carapace tissue. Melanic Daphnia are well known within the D.pulex group, primarily from arctic habitats (e.g. Brooks, 1957; Hebert and McWalter, 1983; Weider et al., 1987; Hobaek et al., 1993), and within the D.longispina group, which is known from both alpine and arctic habitats (Hobaek and Wolf, 1991; Taylor and Hebert, 1994).

The geographical distribution and local occurrence of melanic species and clones indicate that pigmented lineages are highly vulnerable to predation by fish, due to their visual conspicuousness relative to unpigmented variants. This assumption is well founded in theory and empirical observations on visual selection in fish (Zaret, 1980). Apart from body size alone, any colour that increases visibility is generally expected to increase the selectivity of a visually feeding fish for a particular food item. Examples include pigmented eggs, eyes (Zaret, 1972) and ephippia (Mellors, 1975) in cladocerans, and carotenoid pigments in copepods (Hairston, 1979; Luecke and O'Brien, 1981). Nonetheless, the assumption of increased vulnerability of melanic Daphnia to fish predation remains untested in a field situation. This is probably due to the fact that co-occurrence of melanic Daphnia with fish is uncommon.

Here we present the results of a field study that provided the opportunity to

Downloaded from https://academic.oup.com/plankt/article/18/11/2113/1394344 by guest on 21 April 2021

(2)

H&egrov, A.Hobsek and J.H.L'Ab&-Land

document clear selectivity for the pigmented individuals of a Daphnia cf.

longispina population, superimposed on normal size selectivity. The study was conducted in a shallow, clear-water alpine lake with virtually no littoral vegeta- tion. No temperature stratification occurred and Secchi transparency exceeded lake depth. Hence, the abiotic conditions of the lake provided no refuge from fish predation for the zooplankton population.

Method

Lake Bjoraesfjorden (17 km2 surface area) is situated at 1223 m a.s.1., on the Hardangervidda mountain plateau (60°10'N, 7°41'E) in Norway. Most of this clear-water lake is shallower than 6 m and a Secchi disk was easily visible at this depth during the study period in August. Water temperature was 8.2°C through- out the water column. The lake hosts a relatively low-numbered population of large-sized brown trout {Salmo trutta L.), which has been harvested for centuries.

The yield from commercial net catches was estimated at 1.53 kg ha*1 in 1989 (Tysse and Garnas, 1990), and the average length of commercially netted fish was 34 and 38 cm for stocked and wild trout, respectively (L'Ab6e-Lund and Szegrov, 1991).

The data were collected during test fishing in August 1989, -40 days after ice- break. Fishing methods and results on fish stock composition were presented in L'Ab6e-Lund and Saegrov (1991). Zooplankton were sampled by vertical and horizontal net hauls (diameter of net 30 cm, mesh size 90 u.m). Plankton samples were preserved in ethanol, as were samples of fish food contents. In addition to Daphnia cf. longispina, the crustacean zooplankton community included the cladocerans Bosmina longispina and Sida crystallina, the copepods Cyclops scutifer, Mixodiaptomus laciniatus, and Heterocope saliens.

From the net hauls, we measured total length (TL; from base of shell spine to top of head shield) and carapace length (CL; from base of shell spine to base of antennal insertion) in 200 Daphnia individuals by means of a dissecting micro- scope fitted with an ocular micrometer. The linear least-squares regression of TL onCL

TL = 1.216CL + 0.097 (N = 198, R2 = 0.98, P < 0.001)

was then used to estimate TL in damaged Daphnia individuals in the stomach contents. A total of 101 fish stomachs were examined. Up to 20 Daphnia indi- viduals per stomach were measured. Stomach contents were subsampled as necessary. Each Daphnia in the plankton and stomach samples was scored as transparent, intermediate or melanic, based on its appearance under the micro- scope.

Results

Melanic Daphnia constituted 73% of total Daphnia numbers in net hauls, while transparents made up a minor part (6%; Table I). The difference between these 2114

(3)

Vulnerability of melanic Daphnia to brown trout

Table L Number, frequency (%) and lengths (arithmetic means, mm ± 95% confidence limits and ranges) of three pigment categories of D. c£ longispina in net hauls and brown trout stomachs from the alpine Lake Bjornesfjorden

Pigment category

Transparent Intermediate Melanic Total

Net n

11 44 145 200

hauls Frequency

5.5 22.0 72.5 100.0

Mean 1.82 ± 031 1.87 ±0.17 2.05 ± 0.06 2.00 ± 0.04

Range 1.27-2.55 0.90-2.85 0.86-2.66 0.86-2.85

Fish n

7 73 593 673

stomachs Frequency

1.0 10.9 88.1 100.0

Mean 2.49 ± 0.21 2.39 ± 0.07 2.32 ± 0.02 2.33 ± 0.01

Range 2.02-2.70 1.61-330 1.50-3.07 1.50-330

two categories was even more distinct in the brown trout stomachs with 88 and 1%, respectively. Thus, brown trout exerted a highly significant selection on fully melanic individuals (G-test, P < 0.001).

Within each pigmentation category, Daphnia from the net hauls demonstrated a larger range in size than in the fish stomachs. In the lake sample, there was a bimodality in the length distribution, indicating that at least two cohorts or age groups were present (Figure 1). Analysis of variance (ANOVA) of Daphnia length from the net hauls showed that a significant degree of heterogenity existed among the three groups of pigmentation {F2t\<^ = 3.76, P < 0.05). This was due to a larger body size in melanic individuals (Table I). The length of Daphnia from the fish stomachs also showed a significant degree of heterogenity among the three categories of pigmentation (F267o = 3.63, P < 0.05). However, in this case,

o

IT

10

0 -

10 :

20 :

Lake

Stomachs

1.0 1.5 2.0 2.5 3.0 Daphnia length (mm)

3.5

Fig. L Length distribution (%) of transparent (open bars), intermediate (cross-hatched) and melanic (dark) D. cf. longispina in net hauls (upper, n = 200) and brown trout stomachs (lower, n = 673), Lake Bjornesfjorden, 1989. The number of individuals within each category is given in Table I.

(4)

ILSsegrov, AJiobsek and J.HX'Ab£e-Lund

the transparent individuals were larger than the other groups. Irrespective of pigmentation, brown trout selected the larger individuals (ANOVA, P < 0.01 for all groups; Table I, Figure 1).

Discussion

Previous studies on population genetics in alpine populations of the D.longispina complex have revealed that melanic populations constitute a separate genetic entity, which probably is a distinct species (Hobaek and Wolf, 1991). Moreover, melanic and transparent lineages were never found to co-exist, probably because of competitive exclusion and differences in UV tolerance. However, melanic populations may contain a number of apparently transparent animals, which have recently moulted. Since cuticular pigments are lost with each exuvium, they have to be replenished in the newly formed exoskeleton. This process takes some time (at least 1 day at these temperatures; A.Hobaek, unpublished observation).

Genetic analyses and direct observations confirm that such animals are in fact melanic (Hobaek and Wolf, 1991; Hobaek, unpublished observation). The propor- tion of apparently transparent individuals depends directly on moulting frequency, and ultimately on food availability and temperature. Hence, the occurrence of variability in pigmentation level in this Daphnia population is readily explained by a time lag in the deposition of melanin. In the absence of genetic analyses, however, we cannot rule out the co-existence of clones varying in pigmentation level in Lake Bjornesfjorden. Morphologically, the population appears homogenous. Although the population undoubtedly belongs within the D.longispina complex, its relation to established taxa within this group remains uncertain.

Brown trout in Lake Bjornesfjorden fed chiefly on Gammarus lacustris and Lepidurus arcticus, which together made up 96% in dry weight of the stomach contents (L'Ab6e-Lund and Saegrov, 1991). These prey were 50-100 times larger (milligrams dry weight) than D. cf. longispina, which made up 1.5% of total stomach contents. However, Daphnia were found in 20% of the brown trout stomachs with a mean number of 325 in each, demonstrating that it is an attract- ive prey. The presence of melanic D. cf. longispina, despite predation from plank- tivorous brown trout, in Lake Bjornesfjorden is most likely due to an overall low predation pressure from fish. Recruitment limitations, as well as regular harvest- ing, contribute to maintain a low fish stock in this lake (L'Ab6e-Lund and Sajgrov, 1991).

A preference for larger prey is expected for brown trout (Langeland, 1978). In a recent study, Langeland and N0st (1995) found this species capable of catching Daphnia galeata as small as 1.1 mm, and B.longispina down to 0.4 mm. Thus, almost the entire Daphnia population in Lake Bjornesfjorden is vulnerable to predation (Figure 1). The smallest individual eaten was 1.5 mm long. The size component of selectivity was most pronounced for transparent, and less so for pigmented, Daphnia prey (Table I). Melanic individuals were caught at smaller sizes than transparent ones. Hence, brown trout seem to select pelagic prey mainly by its visibility (or apparent size; O'Brien et al., 1976).

2116

(5)

Vulnerability of metallic Daphnia to brown trout

Planktonic Daphnia populations often undertake diurnal vertical migrations to avoid visual predation (Lampert, 1989). Vertical migrations have also been observed in alpine populations (Hessen, 1993), and provide protection from UV radiation as well as fish predation. In shallow Lake Bjornesfjorden, there is no dark refuge in the depths, suggesting that diurnal migration is unimportant. In fact, numerous darkly pigmented animals could be observed by the naked eye at the surface even around noon. This is probably an important prerequisite for the persistence of a pigmented population, since if a depth refuge was available, unpigmented Daphnia clones might well outcompete the melanic lineage(s).

The adaptive value of melanin pigments in Daphnia is fairly well documented (Hebert and Emery, 1990; Hobaek and Wolf, 1991; Hessen, 1994). As previously assumed, this adaptation does indeed render the animals highly vulnerable to fish predation. The co-occurrence of fish and a pigmented Daphnia population in Lake Bjornesfjorden apparently depends on a balance between selection for pigment protection from UV irradiation and the intensity of fish predation. In this shallow lake, there is no spatial escape from either of these strong selective forces. If this is correct, we would expect to see a negative relationship between the densities of brown trout and melanic Daphnia in otherwise comparable lakes, and changes in the fish stock should be followed by changes in melanic Daphnia density.

Acknowledgements

We are indebted to Gudrun Bakkerud and the late Metter Matre for sorting and measuring zooplankton. Thanks are also due to Harald Lura for help during the field work and for suggesting improvements to the manuscript. This study was supported by the Norwegian Research Council via the research program 'Dynam- ics and enhancement of freshwater fish populations'.

References

BrooksJ.L. (1957) The systematics of North American Daphnia. Mem. Conn. Acad, Arts Sci.,13,1-180.

Hairston,N.G.,Jr (1979) The adaptive significance of color polymorphism in two species of Diaptomus (Copepoda). Limnol. Oceanogr., 24,38-44.

Hebert^P.D.N. and Emery.CJ. (1990) The adaptive significance of culticular pigmentation in Daphnia.

Fund Ecol., 4,703-710.

Hebert,P.D.N. and McWalter,D.B. (1983) Cuticular pigmentation in arctic Daphnia: adaptive diversi- fication of asexual lineages. Am. Nat., Y22,286-291.

Hessen,D.O. (1993) DNA-damage and pigmentation in alpine and Arctic zooplankton as bioindi- cators of UV-radiation. Verh. Int. Ver. Limnol., 25,482-486.

HessenJD.O. (1994) Daphnia responses to UV-light. Arch. HydrobioL Beih. Ergebn. Limnol., 43, 185-195.

HessenJD.O. and S0rensen,K. (1990) Photoprotective pigmentation in alpine zooplankton popu- lations. Aqua Fenn., 20,165-170.

Hobsek,A. and Wolf^H.G. (1991) Ecological genetics of Norwegian Daphnia. II. Distribution of Daphnia longispina in relation to short-wave radiation and water colour. Hydrobiologia, 225, 229-243.

Hobaek.A., WeiderJLJ. and WolfJi.G. (1993) Ecological genetics of Norwegian Daphnia. III. Clonal richness in an Arctic apomictic complex. Heredity, 71,323-330.

L'AWe-Lund J.H. and Sa:grov,H. (1991) Resource use, growth and effects of stocking in alpine brown trout, Salmo trutta L. Aquacull Fish. Manage., 22,519-526.

(6)

UScgrov, A.Hobek and J.H-LAb^e-Lund

Lampert.W. (1989) The adaptive significance of diel vertical migration of zooplankton. Funct Ecol., 3,21-27.

Langeland,A. (1978) Effect of fish (Salvelinus alpmus, arctic char) predation on the zooplankton in ten Norwegian lakes. Verh. Int. Ver. Limnol., 20,2065-2069.

Langeland,A. and N0st,T. (1995) Gill raker structure and selective predation on zooplankton by par- ticulate feeding fish. / Fish. Biot., 47,719-732.

Luecke.C. and O'Brien.WJ. (1981) Phototoxicity and fish predation: Selective forces in color morphs in Heterocope. Limnol. Oceanogr., 26,454-460.

Mellors,W.K. (1975) Selective feeding on ephippial Daphnia and the resistance of ephippial eggs to digestion. Ecology, 56,974-980.

O'Brien.WJ., Slade,N.A. and Vinyard.G.L. (1976) Apparent size as the determinant of prey selection by bluegiU sunfish (Lepomis macrochirus). Ecology, 57,1304-1310.

TaylorJDJ. and Hebert.P.D.N. (1994) Genetic assessment of species boundaries in the North Ameri- can Daphnia longispina complex (Crustacea: Daphniidae). ZooL J. Linn. Soc., 110,27-40.

Tysse^A. and GamaXE. (1990) Fiskeribiologisk unders0kjing i Langesj0en og Bjomesfjorden Nore og Uvdal Kommune 1989. Fylkesmannen i Buskerud, Milj0vemavdeUngen. Rapport nr. 11, pp. 1-48 (in Norwegian).

Weider.LJ., Beaton>lJ. and Hebert.P.D.N. (1987) Qonal diversity in high-arctic populations of Daphnia pulex, a polyploid apomictic complex. Evolution, 41,1335-1346.

Zaret.T.M. (1972) Predators, invisible prey, and the nature of polymorphism in the Cladocera (Class Crustacea). Limnol Oceanogr., 17,171-184.

Zaret,T.M. (1980) Predation and Freshwater Communities. Yale University Press, New Haven and London, 187 pp.

Received on April 16,1996; accepted on July 5,1996